AR technology is known which enhances the real world observed by a person, by adding visual information such as a virtual object in a real space. According to AR technology, a user can be made to perceive a virtual object (hereinafter, called an “AR object”) so as if it is present in a real space. A head mounted display, used by a person wearing it on his or her head, a small-sized information terminal such as a head-up display, a smartphone or a tablet, a navigation system, a game device or the like can be included as a display apparatus which makes a user visually recognize an AR object at the same time as an image of a real space. By controlling a binocular parallax, a convergence of both eyes, and a focal length in these display apparatus, an AR object can be made to be stereoscopically viewed. Further, by performing a control which changes the drawing of an AR object corresponding to a shadow, a viewpoint position, or a change in a visual line direction, a stereoscopic feeling of the AR object can be produced.
A dialogue system can also be considered in which a person performs an operation to an AR object by a hand or a finger. However, since an AR object is a virtual object not actually present, a sense of touch is not obtained, even if a person performs a contacting or pressing operation, and so there will be a problem such as an operation by a user being difficult to understand.
For example, an information processing apparatus has been proposed which performs feedback of an operation by stereoscopically displaying a particle, when detecting that a hand of a user has entered into a space region detected by an operation on the space (for example, refer to PTL 1). According to such an information processing apparatus, a user can visually recognize that his or her hand has entered into a space region capable of detecting an operation. However, since visual feedback such as a display of a particle is not able to be given at the time when not entering into a space region capable of detecting, it will be difficult to obtain a specific position relationship or depth information such as whether the hand of the user himself or herself is in front or behind the space region, or whether the hand of the user himself or herself is close to or far from the space region.